[FFmpeg-cvslog] x86/vf_w3fdif: use aligned loads in w3fdif_complex_high

[James Almer]

ffmpeg | branch: master | James Almer <jamrial at gmail.com> | Tue Oct 27 01:49:22 2015 -0300| [d897d4c12de83e6be9e5d081a024ad5f5c548b34] | committer: James Almer

x86/vf_w3fdif: use aligned loads in w3fdif_complex_high

Found-by: Ronald S. Bultje <rsbultje at gmail.com>
Signed-off-by: James Almer <jamrial at gmail.com>

> http://git.videolan.org/gitweb.cgi/ffmpeg.git/?a=commit;h=d897d4c12de83e6be9e5d081a024ad5f5c548b34
---

 libavfilter/x86/vf_w3fdif.asm |    6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

diff --git a/libavfilter/x86/vf_w3fdif.asm b/libavfilter/x86/vf_w3fdif.asm
index fb33f41..c3c73ea 100644
--- a/libavfilter/x86/vf_w3fdif.asm
+++ b/libavfilter/x86/vf_w3fdif.asm
@@ -216,10 +216,8 @@ cglobal w3fdif_complex_high, 5, 13, 10, 0, work_line, in_lines_cur0, in_lines_ad
     pmaddwd                                m9, m4
     paddd                                  m5, m8
     paddd                                  m6, m9
-    movu                                   m8, [work_lineq+offsetq*4]
-    movu                                   m9, [work_lineq+offsetq*4+mmsize]
-    paddd                                  m5, m8
-    paddd                                  m6, m9
+    paddd                                  m5, [work_lineq+offsetq*4]
+    paddd                                  m6, [work_lineq+offsetq*4+mmsize]
     mova               [work_lineq+offsetq*4], m5
     mova        [work_lineq+offsetq*4+mmsize], m6
     add                               offsetq, mmsize/2